Practical Data Science and Artificial Intelligence Course

Health monitoring for Plants using Practical Transfer Learning on AWS

Learn to capture the health of plants using the image of the leaf. You will learn to develop an Image processing Model using deep learning algorithms. Develop a pipeline to capture the new images fed into the NoSQL database. The ML/AI model will be developed, trained, and deployed on the AWS cloud platform. You shall explore the Sagemaker platform of AWS cloud.

Hugging Face models for NLP Generative Chatbot on AWS

Learn to use AWS services to develop a generative chatbot. You will learn the application of NLP models on AWS cloud. Design and Develop an NLP pipeline to train the model on Text corpus stored in PostgreSQL DB.

Practical Real-time Prediction and Analysis on AWS

Learn to train and develop End to End ML models on the data stored on the s3 bucket and then deploy it with a click of a button using the Autopilot feature on AWS Sagemaker service.

AutoML frameworks for Practical ML project implementation

Implement an ML workflow with automated libraries like AutoEDA, Hyperparameter optimization techniques, and AutoML libraries. Benchmark the best solution by developing multiple models to compare the performances and finally choose the optimal solution for the business problems.

AutoML for Predictive Maintenance

Use AutoML libraries to develop an ML model to perform preventive maintenance Manufacturing sector. Learn to ingest data generated from IoT sensors.

AutoML TPOT for End to End Machine Learning Application

Automate the ML pipeline with tree-based AutoML technique Tpot. Learn to tune the AutoML parameters with the help of Hyperparameters. Export the Best pipeline identified by the AutoML TPOT library for deployment.

Standardized Machine Learning Workflow with Data Versioning

With the help of open-source MLOps platforms, learn to deploy a ML pipeline. Understand the data drift and model drift with the help of data validation strategies of MLops tools.

Machine Learning Algorithms from Scratch using MLOps Strategies

With the help of open-source MLOps platforms, learn to deploy an ML pipeline. Understand the data drift and model drift with the help of data validation strategies of MLops tools.

ML Pipeline using Kubeflow with End-to-End deployment and Monitoring

Learn to set up the Machine learning toolkit for Kubernetes (Kubeflow) on cloud services. Develop End to End ML models with Kubeflow pipelines with deployment.

Image Based Problem

Text Based Problem

Structured Dataset

Image-based Product Quality Evaluator using Azure Custom Vision Services

Recommendation Engine for Restaurants to help Serve Customers with Personalized choices

Practical ML Models Using Azure Serverless Architectures

Leveraging Google Cloud IoT and AI/ML services for Practical ML Applications for Live Streaming Data

Ingest Live Streaming data generated by IoT sensors using the pub-sub system on GCP. Use the data warehouse tool BigQuery for storing the data. Create a pipeline to work with the data stored in Bigquery and perform predictive analytics solutions.

Foreign Language Hand Written Address Mapping using NLP Cloud services

Sequential Pattern Mining Using Google Cloud Platform (GCP)

Distributed Computing using Spark for Network Optimization

Learn to work with very large-sized geospatial data with the help of big data distributed framework services. Understand the true power of Apache Spark in processing the large amount of unstructured data using the GraphX library on the Hadoop Cluster.

Machine Learning Models for Predictive Analytics with SparkML

ML models require continuous training for maintaining the expected results of the business. Along with accuracy for ML models, performance is also an essential metric. Training the ML models on big data will be a very time-consuming activity. To address the high latency problem, Apache SparkML is used on the distributed platform for in-memory cluster computing to try and speed up the ML training pipeline by 100x.

Fraud Detection in Procurement on BigData using Practical tools

Design data pipelines to ingest the data from multiple tables. Preprocess the data for model training. Use big data tools for developing Data science solutions for the retail sector to address procurement fraud.